AI firms draft self-governance plans as federal oversight remains limited

Leading artificial intelligence companies are constructing a voluntary framework to oversee their technology and redistribute wealth, operating in a vacuum where the federal government has stepped back from active regulation. With Washington largely hands-off, Anthropic, OpenAI, and other major players are self-regulating their development while designing distinct mechanisms to share the economic gains of the industry with the public.

President Trump has endorsed this basic framework, allowing these companies to regulate themselves and potentially one another outside of formal government monitoring. This voluntary approach relies on public trust in entities that rival the scale of industrialized nations, even as their leaders warn of catastrophic risks and profound economic disruption. The race to create human-surpassing superintelligence continues, with Anthropic, OpenAI, Google, Elon Musk's XSI, Meta, and Nvidia all pursuing different versions of this self-governance model.

Three competing visions for sharing the spoils of the AI boom have emerged, each offering a different answer to who should control the resulting wealth. Anthropic has pledged 5% of its shares to a new public benefit fund on the eve of its initial public offering. At a hypothetical valuation of $2 trillion, this stake would be worth $100 billion, exceeding the current endowment of the Gates Foundation. Including earlier pledges from founders and employees, roughly 20% of Anthropic's shares are earmarked for the public good.

Elon Musk has proposed a direct distribution model, urging Anthropic to give shares directly to the public. On Monday, he pledged to provide every American child with a share of SpaceX next year, a giveaway valued at $10.5 billion at current prices. Musk described this as the "first of many," though he provided no details on how the distribution would be executed. Meanwhile, OpenAI CEO Sam Altman has floated giving the U.S. government a 5% stake in the company through a sovereign wealth fund. Altman envisions an Alaska-style fund that could distribute investment returns directly to Americans, though this plan remains preliminary and would likely require congressional approval.

The feasibility of these plans depends on extraordinary growth projections and billionaire follow-through. Anthropic's ambitions rest on internal estimates of 17% GDP growth by 2030, while Musk has a history of making promises that fail to materialize. Additionally, Anthropic's roots in effective altruism, a philosophy distrusted by the Trump administration and its allies, raise questions about who defines the public good. No industry consensus exists on how best to redistribute expected wealth, but Trump Accounts are emerging as a favored vehicle for free-market leaders.

Billionaire investor Brad Gerstner is championing broad-based ownership through these tax-advantaged funds for children. In July, SpaceX President Gwynne Shotwell and her husband pledged roughly $320 million in company shares to 2 million children via Trump Accounts, offering a potential blueprint for distributing larger portions of Musk's fortune.

Federal authority remains a complicating factor in this experiment in self-governance. Following revelations that Anthropic's system sent an incorrect murder alert to Philadelphia police and engaged improperly with government sites, the White House mandated that AI firms report security breaches and fix any resulting damage. The Super Intelligence Force stated that this reporting and repair protocol is mandatory, asserting federal power just weeks after the president approved a voluntary industry oversight system.